The mammalian olfactory system is able to solve two important problems in chemistry. Specifically, it discriminates between closely related odorant chemicals, but it also detects odorants of vastly different chemical structure. The responses of the olfactory bulb to a small number of relatively similar odorants have suggested some potentially fundamental rules for odor coding by the olfactory system. However, the measured responses constitute only a meager sampling of the large number of receptors, ligands, and perceived odors that are involved in normal olfactory function. We are now in a position to test the generality of current hypotheses by assessing glomerular responsesto 355 odorants by using a 2-deoxyglucose uptake technique that affords access to the entire structure of the olfactory bulb. These odorants include sets of molecules that differ incrementally along a number of as yet untested molecular dimensions. At the same time, the odorants also include molecules differing very greatly in chemical structure, thereby providing a sample more commensurate with the broad capabilities of the olfactory system. In addition to testing hypotheses that have been advanced previously, this large and varied odorant set should give ample opportunity to uncover novel aspects of odor coding. Finally, we have developed a method to place all of the glomerular activity data in a database that we can use to assess quantitative similarities and differences in patterns either across the entire glomerular layer, or in smaller regions of that lamina. This approach should allow us to extract information about the structural features in odorant molecules that are critical for stimulating specific glomerular areas. The approach also should allow us to make additional quantitative predictions regarding similarities in odor perceptions. We plan to use this informatics approach to extract novel hypotheses regarding the relationships between odorants, glomerular activity and odor perception.